ROSA 205 



4. Kennet. Catmore, the only subcristate form seen, W. M. Rogers, 



1886. 



5. Loddon. Waltham. Stubbing's Heath. Near Ashley Hill. 

 Var, CRASsiFOLiA (Wallm. in Liljebl.Utkast. Svensk Flora, 268 ^1816). 

 B. coi-iifolia ;Fries, Nov. Fl. Suec. ed. i, 33, 1814). Syme, E. B. 221, 



472. Fl. Oxf. 109. 



2. Ock, Near Tubney. 5. Loddon. Waltham. 



Var. Watsoni (Baker, in Naturalist, i. (1864) 98, as a species . 

 Syme, E. B. iii. 221, Fl. Oxf. 109. 



1. Isis. Near Cumnor. 



Var. suBCRisTATA (Baker, 1. c. 97, as a species). 

 Syme, E. B. iii. 222. Fl. Oxf. 109. 



1. Isis. Near Appleton. 



B. gJauca is found in Bucks, near Brill, in Oxfordshire and Surrey. 

 The Subcristate group is poorly represented by individuals in Berk- 

 shire — probably not one in five thousand. 



The usual custom of grouping imder i?. canina all the allied varieties of 

 the Dog Rose is here departed from. For some j'ears past our method of 

 studying the variations of the Dog Rose has been based on a very artificial 

 system. It is possible that by separating the unwieldy species of ff. canina, 

 Linn., into groups, we may obtain better results ; at any rate, we are brought 

 more into accord with the system adopted by M. Crepin ; but it must be 

 borne in mind that finality is by no means stated to have been here attained, 

 nor is absolute specific distinction for the groups adopted claimed. It may 

 be tirged witb much force that some of the forms grouped under one or the 

 other head may differ more from one of the members of its own group than 

 it does from a member of another. This much may be said, that M. Crepin 

 has kindly examined my specimens, which had previously been named h\ 

 our best British experts, and the arrangement is based upon the results of 

 his examination, although I do not know whether he agrees with the nomen- 

 clature. 



R. stylosa, Desv. Journ. de Bot. ii. (1809) 317. 



Top. Bot. 152. Syme, E. B. iii. 230, t. 475. Nyman, 231. Fl. Oxf. 104. 



Native. Septal. Hedges and wood borders. Local. Bare in the north, 



and probablj' absent from a large portion of the central chalk 



downs, not unfrequent on the London Clay in the south of the 



county. Shrub. May-June. 



First record (for the aggregate species), B. sysfyla, Bast. At Donnington 



Castle. Mr. Bicheno in Linn. Soc. Trans. 1818, in Smitlis Eng. Fl. ii. 



396, 1824. 



Probably these refer to the variety sysfyla, rather than to true stylosa. 



5. Loddon. A plant very near to true B. stylosa was found near 



Maidenhead. 

 Var. SYSTYLA (Batard, Fl. Maine-et-Loire Suppl. 31, as a species), 

 Baker in Linn. Soc. I.e. p. 239. 



2. Ock. In Bagley Wood, Sister Jane Frances. 



